Carefully limited neovascularization is an important component of several normal physiological processes, including wound repair. In a number of disease this process of new blood vessel growth escapes from normal physiological controls. It contributes to the pathology of many inflammatory diseases and in cancer the progressive growth of most solid tumors depends on a continuously expanding network of new blood vessels. Recently we have identified a new inhibitor of this process of angiogenesis whose production in hamster cells is dependent on the presence in these cells of an active tumor suppressor gene. This proposal is designed to continue this work and to determine. (i) how the suppressor gene supports the elaboration of the inhibitor by testing where it is in the cascade of events from transcription to secretion that cells lacking a suppressor gene are deficient, and what molecules and mechanisms are involved in linking the suppressor gene to the non-angiogenic, and hence less tumorigenic, phenotype is supports. (ii) how the purified inhibitor acts to block angiogenesis by testing the sensitivity of specific functions of hamster and human endothelial to our inhibitor and by testing peptide fragment of the inhibitor itself to define active domains and hopefully to develop a peptide able to block angiogenesis. (iii) whether or not loss of a similar inhibitor of angiogenesis plays a role in human tumor development by testing cell lines from human tumors that have lost the human chromosome carrying the homologue of the hamster suppressor gene for loss of the inhibitor and, if loss is found, by replacing the inhibitor and testing the effect of this on the tumor cell malignant phenotype. At the genetic level loss of multiple suppressor genes is vital to the development of solid human tumors. These experiments should enable us to identify molecules that link the genetic loss of suppressor gene to the phenotype of angiogenesis, known to be vital to the progressive growth of such tumor.